Protein interfaces are of biomedical relevance not only as drug targets but also in cellular aggregates. Cells use liquid-liquid phase transitions involving proteins and other macromolecules to locally sequester components and thereby regulate many processes. The dense liquid like phases of non-enveloped organelles or cellular bodies are a way to change the cytosolic activity of the aggregated species. We will study the mechanisms of protein solubility and phase separations as an organizing principle to characterize protein aggregation and recognition. We will consider the contributions of each of the amino acids to the formation of protein-protein interfaces in various states of aggregation from dimers to phase separated systems. Theoretical analyses of large scale computer simulations will be used to compare with experimental data. The change in the free energy of solution with respect to concentration of proteins and peptides has been difficult to compute. We will make use of our new computational tools to study concentration driven phenomena for peptides and proteins. We can calculate the solubility limits and analyze the change in solution free energy along the aggregation/nucleation coordinate. We hypothesize there is a relationship among protein solution states involved in aggregation and those in recognition between proteins. Solubility is a stringent test of the balance of solvation and protein interactions in current force fields. By analyzing the mechanism in terms of contributions of the solubility of each of the amino acids to the aggregation, we will build more reliable and accurate models of the nonpolar van der Waals solvation components of the mechanism using our new perturbative analysis methods. This is not a protein interface prediction project, but rather we seek to understand the relationship between how any interface, ordered or disordered, is made and the components of the underlying free energy surface driving solubility.